There is a strong influence of gravity on the lung. The lung deforms under its own weight, stretching some regions and compressing others, affecting ventilation and blood flow. The deformation is strongly posture dependent because the thorax (the container shape) is not symmetrical. This asymmetry can be envisioned as a wedge, with the lung hanging from the shorter side in supine posture and the longer side in prone. The effects of asymmetry and gravity are additive in supine posture, but oppose one another in prone. Goal/Significance Early adoption of prone ventilation in severe Acute Respiratory Distress Syndrome (ARDS) markedly improves survival. This is especially true if arterial partial pressure of CO2 and dead space are decreased by prone posture, implying that this intervention recruits blood flow to match regions of well-ventilated lung. Our goal is to understand the mechanisms by which lung deformation and thoracic shape interact to affect pulmonary blood flow and gas exchange. Since prone ventilation greatly complicates patient care, understanding posture effects on blood flow distribution is critically important. This information will be to develop patient-specific metrics to identify which ARDS patients might benefit most from prone ventilation. This work may have other implications for patient management including fluid balance and inotropic therapy. Innovation Can the effects of prone posture on pulmonary blood flow and gas exchange be predicted from regional lung density distribution measured supine? To answer this question we have developed a suite of functional magnetic resonance imaging techniques that allow regional quantification of lung proton density, alveolar ventilation, and perfusion. This allows the evaluation of how container shape and lung deformation alter ventilation-perfusion (VA/Q) matching and regional dead space under clinically relevant conditions. We propose to combine our sophisticated imaging techniques with person-specific modeling of tissue deformation in collaboration with Dr. Merryn Tawhai (University of Auckland) to calculate regional tissue deformation and trans-pulmonary pressure gradients and evaluate the effect on local blood flow and VA/Qmatching. Approach we will test the hypothesis that regions of highly stretched (high local trans-pulmonary pressure) lung corresponding to regions of high V /Q ratio in supine posture will be reduced in prone posture. A resulting in more uniform V /Q matching. In normal subjects we will use data acquired in supine posture A combined with person-specific modeling to predict the effects on blood flow and VA/Q matching in prone posture, and test our predictions against measurements made in prone posture. This information will be used to develop biomarkers for optimal blood flow based on density distribution and thoracic shape (and thus measureable with routine clinical CT) that predict improvement with prone posture. This will be tested in collaboration with Dr. Harm Bogaard (VU Univ. Amsterdam) in a retrospective study of patients with ARDS who have undergone prone ventilation.